In order to reduce maintenance cost of high temperature components in aged thermal power plants, it is necessary to improve accuracy of remaining life assessment methods. High temperature components such as steam turbine rotors are operated under creep loading condition where creep voids initiate and grow on grain boundaries. Development of a void growth prediction method is important for reliable maintenance of these components. Purposes of this study are to clarify void growth behavior of a turbine rotor material under creep loading condition, and to develop void growth simulation program that can predict damage extension process in the components under actual operating condition quantitatively. Creep tests have been conducted on a 1Cr-Mo-V forging material and creep damaged specimens were produced by interrupting the tests. From observation of the creep damaged specimen, spherical shape voids initiate and grow up to their length of 2μm on the grain boundary at initial stage of damage, and then these voids change their shape to crack-like to grow until their length reaches around 10μm. Finally, crack-like voids coalesce each other to form one micro crack along grain boundary. It can be concluded that void growth rate is controlled by diffusion and power law creep under constrained condition based on theoretical consideration of void growth mechanism. Through these discussions, a new void growth model was proposed by modifying conventional models. A void growth simulation program was developed by incorporating the void growth model into a personal computer. It was confirmed that void growth process under the creep condition in the experiment was well predicted by the simulation program. The simulation program was also applied to predict creep void growth behavior in an actual steam turbine rotor during operation.
A simulation method of the creep damage progress combining the random fracture resistance model and FEM elastic-creep analysis was proposed and was applied to the evaluation of Type III and Type IV creep damage respectively in the coarse and fine grain heat affected zone (HAZ) in welded joints of 2 1/4Cr-1Mo steel. Firstly following the previous work on the fine grain HAZ, parameters for the coarse grain HAZ were identified for simulating the increase of number density of small defects on the basis of the random fracture resistance modeling. Secondly the simultaneous damage simulation for both fine and coarse grain HAZ was applied to both the welded joint model test and the burst test of elbow. A possibility of predicting the final failure time of welded joint was also discussed on the basis of the effective stress against the final failure surface. The proposed simulation successfully reproduced the increase of the number density of small defects in the fine and the coarse grain HAZ, and could predict the failure time of welded joints.
Creep-induced microstructural evolution of a martensitic stainless steel, JIS-SUS403, subjected to tensile creep at 873 K, was studied by monitoring the shear-wave attenuation and the velocity using electromagnetic acoustic resonance (EMAR). Taken into account was the observation that the contact-less transduction based on the magnetostrictive effect was the key establishing monitoring of microstructural change in the bulk. The attenuation coefficient shows a peak at around 20% and a minimum value at 50% of the creep life, being independent of the applied stress. This novel phenomenon is interpreted as resulting from microstructure changes, especially, dislocations' recovery, which is supported by TEM observations for dislocation structure. Thus, it was possible to use this EMAR assessment method to assess the progress of damage in metals and predict their remaining creep life.
This paper focused on microstructural changes in a single crystal Ni-base superalloy, CMSX-4, induced by external loadings. It was shown by a series of experiments that an application of the monotonic straining at R.T. and the subse-quent heat treatment at high temperature resulted in a directional coarsening of γ/γ' microstructure, similar to “rafted structure” by creep. A critical strain was found to be needed to build up the microstructural change. The coarsening direc-tions were significantly depending on the histories of straining. Not only monotonic but also cyclic loadings at R.T. and elevated temperatures could induce the similar changes in microstructure. These series of experimental works provide us a new method for failure and damage analysis of superalloy components.
This paper studies creep and creep rupture properties of Sn-8Zn-3Bi lead free solder. Static tensile creep and creep rupture tests were carried out using the solder at 313K, 353K and 398K. Equations for predicting the rupture time and creep strain of the solder were proposed by analyzing the experimental data. The equations predicted the experimental rupture times within a factor of two and the experimental creep strains within a factor of 1.25. The creep rupture and creep deformation characteristics of Sn-8Zn-3Bi were compared with those of Sn-37Pb, Sn-95Pb, and Sn-3.5Ag solders. Sn-8Zn-3Bi solder had a superior creep resitance to the two Sn-Pb lead and Sn-Ag lead free solders. The relationship between the creep strain rate and relaxation behavior in creep-fatigue test was discussed for the three solders.
Effects of grain size on creep-fatigue properties of an austenitic stainless steel under multiaxial stress states were investigated. Creep-fatigue tests under uniaxial and torsional loading conditions were conducted on 18Cr-9Ni-3Cu-Nb-N steels which were solution treated at three different temperatures. The average grain sizes of the tested materials were 18, 29 and 76μm depending on the temperature of solution treatment. It was clarified that the creep-fatigue life under the torsional loading increased with decrease of the grain size as well as under the uniaxial loading. Observation results of surface crack growth behavior indicated that the initiation life of cracks of 100∼200μm length was longer in the fine grained steel than that in the coarse grained steel. The grain size had larger influence on the fatigue life under CP (slow-fast) type strain waveforms than that under PP (fast-fast) type strain waveforms in both uniaxial and torsional loading conditions. These experimental results were closely related to the tensile and creep ductility of fine and coarse grained steels. It was also found that the creep-fatigue life evaluation method which was based on the strain range partitioning concept and modified with multiaxiality factor, MF, was effective for all tested steels independent of grain size and loading condition.
Fully reversed strain-controlled low-cycle fatigue tests, so called PP tests, were conducted on thin-walled tubular specimens of SUS304 austenitic stainless steel at 973K in air under proportional and nonproportinal loading conditions. Push-pull straining, cyclic torsion straining and in-phase straining of push-pull and cyclic torsion were selected as proportional loading, and 90deg out-of-phase straining of push-pull and cyclic torsion as nonproportional loading. Experimental data were analyzed by using both the Manson-Coffin equation and the critical plane model proposed by Brown and Miller, and the effects of the multiaxial stress state on the strain range versus life relationship were discussed. As the results, a new parameter was proposed that has a hopeful potential to estimate the effect of nonproportional loading on creep-fatigue life quantitatively based on the strain range partitioning method.
Low cycle fatigue tests at the temperature of 600°C were carried out using a component specimen of 12%-Cr steel, which simulates the blade root and disc joint for turbines. Micro-cracks growth behavior at the joint region of the specimens was investigated to clarify the damage mechanism of the blade-root joints used in high temperature environments and to improve life assessment methods using a finite element analysis for them. Micro-cracks growth behavior similar to the smooth bar specimens was observed in the specimens tested under the conditions of relatively high total strain range. Micro-cracks initiation was observed at the notch region of the specimens in the early stage of life. The crack growth rate increased with surface crack length. Life of the component specimens under this condition was similar to the smooth bar specimens. While, the component specimens tested under the conditions of relatively low total strain range showed a different micro-cracks growth behavior. No cracks were observed at the notch region and some micro-cracks were initiated at the edge of the contact region of the specimens in the early stage of life. Almost no increase of the crack growth rate was observed. Life of the component specimens under this condition was shorter than the smooth bar specimens. This might be attributed to the fretting fatigue at the contact edge and a mean stresses. Life prediction using a normalized crack growth rate (da/dN/ae) was effective for the component specimens over a wide range of the strain range.
This study describes the evaluation of material dependence of multiaxial low cycle fatigue to develop a suitable strain parameter for life prediction under non-proportional loading. Strain controlled multiaxial low cycle fatigue tests using 4 types of proportional and non-proportional strain histories were carried out on SUS316NG and SGV410 hollow cylinder specimens at room temperature. The reduction of low cycle fatigue life due to non-proportional loading was discussed relating to an additional hardening and crack initiation and propagation behaviors, and the applicability of the nonproportional strain proposed by the authors was also discussed. Fatigue lives decreased down to 1/5 by the non-proportional loading depending on strain history for SUS316NG and SGV410 steels, whereas the degree of additional hardening was different, which suggested that another factor like difference of crystal structure should be taken into account for the life prediction under non-proportional loading.
This paper describes the circmferential crack propagation in multiaxial low cycle fatigue loading. Push-pull, reversed torsion and combined push-pull and reversed torsion low cycle fatigue tests were carried out using circumferential notched specimens of type 304 stainless steel at 873K and crack propagation rates were measured by a D. C. potential method. Elastic-plastic finite element analysis was also carried out to evaluate the crack and notch opening displacements and to examine the effect of frictional between cracked faces on Mode III crack. The friction decreased the crack opening displacement and plastic zone size ahead of crack tip. Crack propagation rates in push-pull, reversed torsion and combined push-pull and reversed torsion tests were not correlated with J integral range ΔJE calculated by conventional method and which was discussed in relation to the friction between the fracture faces. The crack opening displacement taking account of the friction effect correlated crack propagation rates in all loading modes within a factor of four band.
The bulk metallic glass (BMG) is highly expected for the new structural material by its various excellent properties such as the high strength, the corrosion resistance and so on. Moreover, it can show the superplastic deformation in a high temperature condition, although it has very little plastic deformation in the room temperature. The excellent properties of BMG are caused by its amorphous structure, but its atomic structure tends to change into crystalline structure across TTT curve through a thermal process in a high temperature. Therefore it is important for an application of BMG to study not only its workability with plastic deformation at high temperature but also its mechanical properties at room temperature after thermoplastic deformation. In this paper, the effect on the strength of Zr55Cu30Al10Ni5 bulk metallic glass at room temperature by heating process or thermoplastic deformation in high temperature is mentioned. This material had the tendency to exceed easily its TTT curve and crystallize by heating beyond 685K in a furnace. And the strength of crystallized material at room temperature drastically decreased, while this material as cast with amorphous structure had the high strength. On the other hand, it was made sure that this material could be sufficiently deformed plastically at 667K (18K lower than 685K), and furthermore, it maintained the high strength similar to Zr55Cu30Al10Ni5 BMG as cast.
In order to investigate yielding in amorphous polymers, shear and tensile tests have been performed in the conventional ranges of strain rate and temperature using amorphous samples. In the serial study, when the experiments on the effect of strain rate on yield stress have been executed at room temperature using amorphous polyethylene terephthalate (PET) samples, the yield mode was discovered to change drastically at a critical strain rate. In this paper, this yield mode transition was investigated at shearing. It was shown that 1.there were two different yield modes, (a) yielding governed mainly by localized shear bands (; LSB mode) and (b) yielding by diffuse slip lines (; DSL mode) and 2.while the LSB mode appeared in the case where the strain rate was high and the temperature was low, the DSL one occurred where the strain rate was low and the temperature was high.
The properties of high fluidity concrete with a combination of the fine powder of slowly cooled blast furnace slag and granulated blast furnace slag were examined. The change of slump flow value with time decreased when a specified ratio or more of slowly cooled blast furnace slag was used. The concrete with various levels of compressive strength can be prepared by changing the mixing ratio of these powders. It has been clarified that close relations exist between adiabatic temperature rise and the unit weight of binder, as well as between autogenous shrinkage and water binder ratio, and that mix proportion design suited to required performance is possible. By the use of this high fluidity concrete, it is possible to suppress both CO2 emission and limestone consumption per unit of concrete, which suggests that this concrete is useful from the viewpoint of reduction in environmental burden.